Updated Optimizely application UI

📘

Note

For a full side-by-side comparison of the Full Stack (legacy) and Feature Experimentation refer to our user documentation page on Migration Examples.

Experiments and Features into one tab

The Experiments and Features tabs have been replaced with one Flags tab. This is the central dashboard for your Flags and their Rules which support targeted deliveries and experiments within an Environment in your project. Everything starts with a Flag, whether running an experiment or just rolling out a feature.

Full Stack - Legacy

Optimizely Feature Experimentation

Flag details

The new Flag Details page is where you can manage a Flag's Ruleset. The Ruleset uses sequential logic for delivering your feature or experiments. Aside from all experiments now requiring a Flag, all other components for creating experiments (Audiences, Metrics and Variations) remain the same as in the Legacy experience.

📘

Note

Having multiple rules per flag is currently in Beta. If you would like to be added to the beta, please reach out to your customer support representative.

You can create multiple Rules for the same Flag, providing flexibility in the experiences you deliver to users. Rules are evaluated sequentially, allowing you to maximize customer experience, given your priorities.

For example, you can prioritize your qualifying traffic to a flag for an experiment and delegate non-qualifying traffic to a default experience or targeted delivery. Because all experiments are based on flags, all winning variations in Feature Experimentation can be rolled out without needing to go through your entire software development lifecycle, including an extra deployment.

Shared variations

Variations and their Flag Variables are now shared across all Rules in a Flag, making it easy to reuse variations across your Rules.

Different results per environment

In Full Stack (legacy), a single experiment exists across all environments and shares one results page. When you QA an experiment, those events get mixed with the live production results. In Feature Experimentation, Experiment Rules are scoped by Environment, so results are scoped by the environment. This eliminates a common pain point wherein events created in QA are mixed with live production results and the need to reset results before running the Experiment in your primary environment. Refer to How existing experiments are migrated for more information.

Reports page

In Full Stack (legacy), experiment results can only be viewed at the experiment level, requiring clicking view results for each experiment and not providing a centralized view of your results.

Feature Experimentation introduces a new Reports tab, where you can access results for your experiments from one view. Reports are displayed for one environment at a time and can be changed with the filter options.